Universal IoT-Enabled Smart Battery Charging Optimization: Leveraging AI, Automata Theory, and Real-Time Data for Enhanced Lifespan and Efficiency
Year: 2025
Type: conference-paper
Abstract: Universal IoT-Enabled Smart Battery Charging Optimization System introduces a novel solution to addressing the widespread problem of battery degradation in IoT-enabled devices such as smartphones, wearables, stylus pens, and wireless earbuds. The system utilizes Artificial Intelligence (AI), Automata Theory, and sensor data in real-time to develop a dynamic context-aware charging strategy that learns from the unique usage patterns and battery status of every device. By applying machine learning models like Kalman Filters, Recurrent Neural Networks (RNNs), and Finite State Machines (FSM), the system charges behaviors smartly, optimizing power consumption, charging efficiency, and battery life. Pushdown Automata (PDA) are used to facilitate non-deterministic state transitions from past charging data, providing more precise predictions of battery wear. The system constantly collects and processes data from IoT sensors with real-time charging cycle and battery health feedback. Initial results indicate that the system greatly mitigates overcharging, enhances battery life, and promotes energy efficiency, providing a green solution to maximize the lifespan of IoT-based devices in various applications. The solution offers a sound framework for further research on smart energy management and green computing.
Source: EPJ Web of Conferences
Authors: Divyansh Arora, Ananya Goswami, Mritunjay Ranjan, Arif Md Sattar
Institutions: Delhi Technological University, Sandip Foundation
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4411430439